© Copyright JASSS

JASSS logo ------

Philosophy and Simulation

DeLanda, Manuel
Continuum Publishing Corporation: London, United Kingdom, 2012
ISBN 978-1441170286 (hb)

Order this book

Reviewed by Martin Neumann
University of Koblenz-Landau

Cover of book The book provides a big picture of self-organized systems from thunderstorms to civilizations. In particular when it reaches the domain of human life the book discusses a lot of simulation models that many readers of JASSS might be familiar with: evolutionary game theory in the Axelrod style, Sugarscape, the EOS model, or the Anasazi model, to mention just a few. So it might be interesting for JASSS readers to learn about a philosopher’s point of view.

However, a few things need to be noted in advance: First, the book is not about philosophy of simulation. The book regards simulation models as laboratories, a point of view which is debated in the scholarly discourse on philosophy of simulation. For instance, it remains unaffected by the work of Dowling (1999), Winsberg (2003), or Humphrey’s (2004) book Extending ourselves, to mention just a few classics. While the book discusses simulation models, examining them is not its final aim. Ultimately, the book is about emergence and uses simulation as a means for its study. However, again it has to be noted that the book does not care much about the scholarly discussion of emergence. Only in the introduction, 19th and early 20th centuries’ account of emergence is briefly mentioned as a flawed approach to philosophy of science. Samuel Alexander’s (1920) famous phrase that emergence “has to be accepted with natural piety” is the point of departure. DeLanda consequently ignores the literature about the re-emergence of emergence theory with the rise of complexity science and the debates of the work of, among others, Kim (1992), Goldstein (1999), or Sawyer (2003, 2005). Finally it has to be noted that many models discussed in the book date back to the mid1990s and only very few are of the last decade. As already Hegel said: Philosophy is the owl in the night.

Taking these warnings into account the book offers fascinating material to read, full of puzzling insights. Throughout the chapters the problem is sketched first and then it is discussed how simulation models help in explaining mechanisms of emergence. After a first sketch of the problem, the book starts studying emergence in physical systems: thunderstorms. Gradients in differences of temperature serve as energy storage which is the central mechanism of how thunderstorms are born and die, studied by equations of mathematical models, complemented by simulations that provide a lab for investigating actual flow of particles. The subsequent chapter on cellular automata shows how the game of life as most prominent example thereof can simulate phase transitions of water particles during the storm.

The book goes on to the next layer of emergent properties in chemical systems. Chemical gradients of concentration of substances in water generate an autocatalytic loop in the prebiotic soup. Recursive functions uncover the fitness landscapes of the chemical processes just before the emergence of life. This is characterized by the capacity of self-replication of RNA that now can be studied by genetic algorithms.

The next layer of emergence is ancient organisms, gelatinous stratum of colonial bacteria in an aqueous environment. These are responsible for the central inventions in the history of life: fermentation, photosynthesis, and respiration. Note that this happened about 3 ½ billion years ago, whereas the evolution of multicellular species is only 600 million years ago.

These are subject of the following chapters on insect intelligence and mammalian memory. Already rather primitive creatures like jellyfish evolved sensors that modulate behavioral responses to stimuli. This is the beginning of some form of internal representation of the environment. While insects evolutionary learning takes place at population level mammals can learn as individuals by an autobiographic memory. Non-linguistic representation can be simulated by increasingly complex neural nets. A review of multi-agent simulation models starts in the following chapter on primates: they establish a social gradient of group solidarity that can be studied by evolutionary game theory.

Finally, three chapters are dedicated to human civilization: stone-age economy, the emergence of language, and ancient states. Exchange of stone artifacts is roughly traced back to 500 000 years, already prior to the emergence of language. Structural conditions can be modeled mathematically by an Edgeworth Box, whereas studying mechanisms of exchange require agent based simulations such as the exchange of sugar and spice in the Surgarscape model. As only scarce data exist how sound has been associated with meaning, the chapter on the evolution of language relies mostly on automata theory for classifying possible languages of increasing complexity. Lastly it is discussed how social stratification emerged from egalitarian Neolithic tribes to organizations that overcome the limitations of personal agency enabled, e.g., in erecting the pyramids of Giza. Status differentiation needs to be ensured by a gradient of legitimacy. Presumably, as this requires magical beliefs, the author suggests BDI agents. However, already simulations that concentrate on material conditions and organizational behavior provide insights.

The book is refreshing to read as the author does not engage in scholarly debates but provides his own view of the world straightforwardly. I’m not sure whether I shall follow in every aspect. In particular with regard to the central philosophical claim of an ontological, emergentist realism I would like to read a scholarly discussion before forming an opinion. However, the book introduces numerous thought stimulating ideas: I was impressed by the unagitated view on simulation complementing mathematical equations. DeLanda separates two elements of an explanation: Mechanism independent elements such as equilibria in economics. These are more of the kind of general properties that can be studied by mathematical equations. However, this element lacks a specification of the mechanisms which bring about the general properties, as for instance, interactions actors. These mechanisms can at best be studied by simulation models which show for instance how a certain equilibrium is reached in concrete interactions.

Throughout the chapters the book provides numerous examples of how simulation enables the investigation of the mechanisms of emergence. DeLanda describes emergent properties by tendencies and capacities that need not be actually realized but form a possibility space that can be studied by simulation. The chapters on empirical cases of emergence provide concrete examples. This is summarized in a theory called assemblage theory as a new start for old debates. This unorthodox philosophy stimulates seeing things from a different perspective that can be recommended also to non-philosophical readers.

* References

ALEXANDER, S. (1920). Space, Time, and Deity, Vol. II. London: MacMillan.

DOWLING, D. (1999). Experimenting on theories. Science in Context 12(2): 261-273.

GOLDSTEIN, J.A. (1999). Emergence as a construct: History and Issues. Emergence 1(1): 49-72.

HUMPHREY, P. (2004). Extending ourselves. Computational science, empiricism, and scientific method. Oxford: Oxford University Press.

KIM, J. (1992). Downward Causation in Emergentism and Nonreductive Physicalism. In: Beckermann, A., Flor, H., Kim, J. (eds.) Emergence or Reduction? Berlin: De Gruyter, pp. 119-138.

SAWYER, K. (2003). The Mechanisms of Emergence. Philosophy of the Social Sciences, 34(2): 260-282.

SAWYER, K. (2005). Social emergence. Societies as complex systems. Cambridge: Cambridge University Press.

WINSBERG, E. (2003). Simulated Experiments: Methodology for a Virtual World. Philosophy of science, 70(1): 105-125.


ButtonReturn to Contents of this issue